Recent MRI studies demonstrate progressive changes in gray matter in temporal and frontal cortices following onset of schizophrenia. Far less, however, is known about white matter abnormalities, particularly fronto-temporal connections, long thought to be abnormal in schizophrenia. Additionally, post-mortem studies indicate that astrocytes and oligodendrocytes may be abnormal. Since glial cells play an important role in neuronal migration and in synaptic function, their dysfunction could lead to reduced neuronal size, reduced levels of synaptic proteins, as well as to abnormalities in neurotransmission and functional dysconnectivity. Of further note, recent genetic studies point to oligodendrocyte and myelin related (OMR) abnormalities. Taken together, these findings indicate that an investigation of white matter may lead to a further understanding of the neuropathology of schizophrenia. The main aim of this project is to investigate evidence for vulnerability to progression of white matter abnormalities in schizophrenia, at various stages of the illness (i.e., before, at first onset, long after illness onset), in order to delineate possible putative markers. Subjects will be: (1) 75 prodrome at risk subjects at study entry and at 1-year follow-up, and for those who convert to a first episode of schizophrenia, also at time of entry into the first episode study; (2) 80 first episode schizophrenia, defined by first hospitalization, at study entry.and followed at 6-mths and 18-mths; (3) 42 patients diagnosed with chronic schizophrenia. Normal controls will be group matched for age, gender, etc., to each cohort. We will use Diffusion Tensor Imaging to evaluate fronto-temporal white matter fiber connections (uncinate fasciculus, cingulate bundle, arcuate fasciculus, and fornix), corpus callosum, anterior commissure, and anterior limb of internal capsule, where we predict decreased anisotropy in chronic schizophrenics>first episode with changes at 18 months>first episode changes at 6 months>prodromes who convert>prodromes who do not convert. We also predict associations with OMR genes, where we will test for association between gene sequence variants (single nucleotide polymorphisms-SNPs) for six OMR genes (NRG1, ErbBS, MAG, CNP, MOG, GRM3). (See also Postmortem Core for isolation of mRNA from 4 OMR genes-ErbB3, MAG, CNP, MOG). This proposal will provide new discoveries relevant to white matter progression in schizophrenia, which will likely lead to new treatment and preventative strategies.